Advisory speed for Intelligent Speed Adaptation in adverse conditions

In this paper, a novel approach to compute advisory speeds to be used in an adaptive Intelligent Speed Adaptation system (ISA) is proposed. This method is designed to be embedded in the vehicles. It estimates an appropriate speed by fusing in real-time the outputs of ego sensors which detect adverse conditions with roadway characteristics transmitted by distant servers. The method presents two major novelties. First, the 85^th percentile of observed speeds (V₈₅) is estimated along a road, this speed profile is considered as a reference speed practised and practicable in ideal conditions for a lonely vehicle. In adverse conditions, this reference speed is modulated in order to account for lowered friction and lowered visibility distance (top-down approach). Second, this method allows us taking into account the potential seriousness of crashes using a generic scenario of accident. Within this scenario, the difference in speed that should be applied in adverse conditions is estimated so that global injury risk is the same as in ideal conditions.